How to scale above clouds limits - Radu Vunvulea
Download as PPTX, PDF0 likes653 views
The document outlines strategies for scaling cloud applications beyond typical limits, emphasizing the importance of understanding limitations and implementing effective failover mechanisms. Key concepts discussed include scaling units, storage capacity, and different failover strategies such as active-active and active-passive configurations. It highlights the necessity of communication channels and the potential challenges associated with scaling solutions.
How to scale above clouds limits - Radu Vunvulea
- 1. How to scale above clouds limits Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia
- 2. Why scalling is important ISTAcon.org 18-19 November, Sofia
- 3. Why scalling is important ISTAcon.org 18-19 November, Sofia 8000m – Cirrus 6000m – Cumulonimbus 4000m – Altostratus 2000m - Nimbostratus
- 4. Understand that scaling and failover is not only cloud provider’s responsibilities, is also our yours ISTAcon.org 18-19 November, Sofia
- 5. Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia RADU VUNVULEA MCTS MCP BANK HOME AUTOMATION MVP IoT AUTOMOTIVE PHARMA LEAN AND AGILE E-COMMERCE WEB&SPA iQuest AZURE JAVASCRIPT VUNVULEARADU.BLOGSPOT.COM MOBILE DOTNET @RaduVunvulea WCF WPF ENTHUSIASTIC
- 6. Agenda ISTAcon.org 18-19 November, Sofia • Identify limitations • Investigate different problems • Overview on different solutions
- 7. Hardware Capabilities Functionality App Design Types of limitations ISTAcon.org 18-19 November, Sofia
- 8. Types of limitations ISTAcon.org 18-19 November, Sofia Hardware Capabilities Functionality App Design
- 9. Types of limitations ISTAcon.org 18-19 November, Sofia Hardware Capabilities Functionality App Design
- 10. Types of limitations ISTAcon.org 18-19 November, Sofia Hardware Capabilities Functionality App Design
- 11. Types of limitations ISTAcon.org 18-19 November, Sofia Hardware Capabilities Functionality App Design
- 12. Storage bandwidth Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia
- 13. Storage bandwidth ISTAcon.org 18-19 November, Sofia Target throughput for single blob • 60MB/s • 500 requests/s Per storage account • 30Gb/s
- 14. Storage bandwidth ISTAcon.org 18-19 November, Sofia Target throughput for single blob • 60MB/s Can we still have 10MB/s for each client?
- 15. Storage bandwidth ISTAcon.org 18-19 November, Sofia Target throughput for single blob • 60MB/s Can we still have 10MB/s for each client? NO
- 16. Storage bandwidth ISTAcon.org 18-19 November, Sofia Solutions • CDNs • Content replication
- 17. Storage bandwidth ISTAcon.org 18-19 November, Sofia Solutions • CDNs • Content replication
- 18. Storage capacity Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia
- 19. Storage capacity ISTAcon.org 18-19 November, Sofia Storage account capacity • 500TB Storage accounts • 100 per subscription
- 20. Storage capacity ISTAcon.org 18-19 November, Sofia Storage account capacity • 500TB What happens in 1 year if we produce 2TB of data per day?
- 21. Storage capacity ISTAcon.org 18-19 November, Sofia Storage account capacity • 500TB What happens in 1 year if we produce 2TB of data per day? Out of space
- 22. Storage capacity ISTAcon.org 18-19 November, Sofia Solution: • Work with multiple storage accounts To be able to do this our application needs to be able to store a list of storage accounts, not only one
- 23. How would you store and manage the configuration of multiple storage accounts? ISTAcon.org 18-19 November, Sofia
- 24. Storage configuration ISTAcon.org 18-19 November, Sofia • CSDEF • SQL Database • Azure Table • Azure Storage
- 25. Storage configuration ISTAcon.org 18-19 November, Sofia • CSDEF • SQL Database • Azure Table • Azure Storage • Blobs • XML Format
- 26. How long we can scale? Adding more and more instances and resources is a good solution. Can we quarantee the quality of our service when we are adding more and more resources? How much we can stretch in this way? ISTAcon.org 18-19 November, Sofia
- 27. Adding more instances ISTAcon.org 18-19 November, Sofia
- 28. ISTAcon.org 18-19 November, Sofia Adding more instances
- 29. Adding more instances ISTAcon.org 18-19 November, Sofia
- 30. Adding more instances ISTAcon.org 18-19 November, Sofia • Soon this will break, why? • The quality of service will not remain the same by adding more and more resources • Managing a large infrastructure will be a nightmare • Our confidence will be lower and lower
- 31. Adding more instances ISTAcon.org 18-19 November, Sofia • Soon this will break, why? • The quality of service will not remain the same by adding more and more resources • Managing a large infrastructure will be a nightmare • Our confidence will be lower and lower • In the end the system will BREAK
- 32. Scaling Units ISTAcon.org 18-19 November, Sofia
- 33. Scaling Unit ISTAcon.org 18-19 November, Sofia • Grouping resources together and defining the quality attributes of that scaling unit • When the limits are hit another scaling unit is added, without adding more resources to the current one(s)
- 34. Scaling Unit ISTAcon.org 18-19 November, Sofia
- 35. Scaling Unit ISTAcon.org 18-19 November, Sofia
- 36. Scaling Unit ISTAcon.org 18-19 November, Sofia
- 37. Holy Cow! This is what I call a major point! ISTAcon.org 18-19 November, Sofia
- 38. Failover mechanism What should we do when we need better SLA? Or a failover mechanism… … to be able to be up and running even during and after a major disaster… ISTAcon.org 18-19 November, Sofia
- 39. Failover mechanism ISTAcon.org 18-19 November, Sofia • One or more services from current region are going down • What should we do?
- 40. Failover mechanism ISTAcon.org 18-19 November, Sofia • Adding resources in another region
- 41. Failover mechanism ISTAcon.org 18-19 November, Sofia • Adding resources in another region • Before • During • After a disaster
- 42. Failover mechanism ISTAcon.org 18-19 November, Sofia • Adding resources in another region • Before, is great. But with extra costs • During, we can lose a part of data • After, is to late. We are not the only one in this situation a disaster
- 43. Active - Passive ISTAcon.org 18-19 November, Sofia
- 44. Active - Passive ISTAcon.org 18-19 November, Sofia • Active is up Only storage is allocated on the passive
- 45. Active - Passive ISTAcon.org 18-19 November, Sofia • Active is up
- 46. Active - Passive ISTAcon.org 18-19 November, Sofia • Active is down
- 47. Active - Active ISTAcon.org 18-19 November, Sofia
- 48. Active - Active ISTAcon.org 18-19 November, Sofia • Active is up 75% of computation power is allocated on passive
- 49. Active - Active ISTAcon.org 18-19 November, Sofia • Active is up 75% of computation power is allocated on passive
- 50. Active - Active ISTAcon.org 18-19 November, Sofia • Active is down
- 51. Active – Active vs Active - Passive ISTAcon.org 18-19 November, Sofia Active-Active • Minimum downtime • Resource reservation • You pay for resources that you don’t use • Data sync can be a nightmare Active-Passive • A few seconds (minutes) downtime • No resources reservation • No extra costs (except storage) • Data sync can be a nightmare
- 52. Do I really need Active-Active? ISTAcon.org 18-19 November, Sofia
- 53. Do I really need Active-Passive? ISTAcon.org 18-19 November, Sofia
- 54. What about communication It is not enough to have a failover mechanism for storage and computing If there is no communication channel between the components then we cannot used them … ISTAcon.org 18-19 November, Sofia
- 55. Communication channel ISTAcon.org 18-19 November, Sofia
- 56. Communication channel ISTAcon.org 18-19 November, Sofia • Active – Active • Active – Passive
- 57. Communication channel ISTAcon.org 18-19 November, Sofia • Active – Passive
- 58. Communication channel ISTAcon.org 18-19 November, Sofia • Active – Passive
- 59. Communication channel ISTAcon.org 18-19 November, Sofia • Active – Active
- 60. Final conclusion Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia
- 61. Conclusion ISTAcon.org 18-19 November, Sofia • We should scale our system in different ways, not only by adding more instances • Scaling Unit is a good way for scaling • Active-Active can be useful but is expensive and can be a nightmare
- 62. Q&A Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia
- 63. Thank you! @RaduVunvulea vunvulearadu.blogspot.com linkedin.com/in/raduvunvulea vunvulear#gmail.com Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia
- 64. Take a look on Scaling Unit concept (Next steps) ISTAcon.org 18-19 November, Sofia
- 65. Thank you! @RaduVunvulea vunvulearadu.blogspot.com linkedin.com/in/raduvunvulea vunvulear#gmail.com Radu Vunvulea | @RaduVunvulea 18-19 November, Sofia
- 67. Major point without a picture And its subtitle ISTAcon.org 18-19 November, Sofia
- 68. One Title to Rule them All One subtitle to find them. Mr Frodo Baggins | The Shire Inc. 18-19 November, Sofia
- 69. • Up to • Six bullets • For best results • Up to • Six bullets • For best results Integer tincidunt libero a ex venenatis cursus. Vivamus accumsan congue tristique. ISTAcon.org 18-19 November, Sofia
- 70. Integer tincidunt libero a ex venenatis cursus. Vivamus accumsan congue tristique. ISTAcon.org 18-19 November, Sofia
- 71. • Up to • Six bullets • For best results • Up to • Six bullets • For best results Integer tincidunt libero a ex venenatis cursus. Vivamus accumsan congue tristique. ISTAcon.org 18-19 November, Sofia
- 72. Integer tincidunt libero a ex venenatis cursus. Vivamus accumsan congue tristique. ISTAcon.org 18-19 November, Sofia
- 73. ISTAcon.org Some icons for use throughout the presentation 18-19 November, Sofia
- 74. Thank you! @TwitterNickname PersonalWebsite.com linkedin.com/in/frazzleddad name@mailprovider.com Mr Frodo Baggins | The Shire Inc. 18-19 November, Sofia